-
Arch. Dis. Childh., 1964, 39, 125.
DIASTEMATOMYELIAA CRITICAL SURVEY OF 24 CASES SUBMITTED TO
LAMINECTOMY*
BY
C. C. MICHAEL JAMES and L. P. LASSMANFrom the Regional
Neutrological Centre, Newcastle General Hospital, and
W. J. Sanderson Orthopaedic Hospital, Newcastle upon Tyne
We have previously described a syndrome associa-ted with spina
bifida occulta where there is acongenital abnormality, extrinsic to
the spinal cord,causing neurological deficit in the lower
limbs,bladder and bowel. We have operated on morethan 70 patients
and of the first 60, 24 haddiastematomyelia. It is our purpose in
thiscommunication to describe the forms of diastemato-myelia found
and to analyse the clinical findings.A few of these patients have
been previously reportedon individually in detail with photographs
taken atoperation (James and Lassman, 1958, 1960, 1962a, b;Lassman
and James, 1963).
Diastematomyelia, which is a bifid state of thespinal cord, is
an intrinsic anomaly and requires notreatment. The spinal cord is
known to be able tofunction normally in such cases but it may be
affectedby associated extrinsic abnormalities that interferewith
conduction, either directly or indirectly. Itdoes not seem to be
sufficiently recognized that thiscondition is indistinguishable
clinically from theother types of lesion that are associated with
thespinal cord in spina bifida occulta and that produce asimilar
pattern of neurological deficit.
In many cases the two spinal cords are containedwithin a single
dural tube without an interveningseptum, but when each spinal cord
has its own duralsheath there is always a septum of bone,
cartilage,fibro-cartilage or fibrous tissue that can prevent
thespinal cord from 'ascending' within the vertebralcanal as the
vertebral column grows in length, or itcan cause pressure laterally
on one or other spinalcord. The results of septal pressure are,
therefore,more likely to be seen in childhood although theyhave
been reported in adult life.
Operation FindingsEvery case had spina bifida of a greater
degree
than a simple split in the spinous process of Sv.l as
* A paper read at a meeting of the British Association of
PaediatricSurgeons in Sheffield, July 1963.
seen on radiography. The spinal cord abnormalitywas found in
close proximity to the laminal defectsbut not always at the same
level. This is particularlynotable in cases with a bone septum. In
these thelaminae continuous with the septum always appearednormal,
and in fact the bone septum is frequentlyunnoticed in plain x-ray
films because of its resem-blance to a spinous process in the
antero-posteriorview. Myelography demonstrates diastematomyeliain
the majority of cases (Gryspeerdt, 1963).
Cases with Separate Dural Tubes. There were 13of these and each
of them had a septum external tothe two median layers of dura. The
septum wasclearly affecting the spinal cord in six cases, in one
bylateral pressure and in the other five by preventingascent. The
dura is also a factor causing injurysince it is the first tissue to
come in contact with theseptum, and where ascent is being prevented
the tubeof dura may have a constricting effect as well as beingthe
point of contact with the caudal junction of thetwo spinal cords
(Fig. 1).The degree of ossification of the septum was not
related to age, for in one child of 10 years the septumwas
fibrous and yet in another child of 2 years it waswell
ossified.
In the operation, it has been found safer and easierto remove
the septum before opening the dura whichacts as a protective
covering for the spinal cord.Although the septum usually occupies
the whole ofthe space between the two dural tubes,
thediastematomyelia within the dura is frequently ofgreater extent
and therefore not so closely adherent.This allows a margin of
safety in manipulation of theinstruments. The median dura between
the spinalcords is excised later when the intrathecal situationis
plainly visible and it can be ensured that nothing isleft that
might interfere with any future need forchange in position of the
spinal cord within thevertebral canal during the course of normal
growth.
In the majority of our cases, the broadest attach-ment of the
septum has been dorsal, and where it was
125
copyright. on M
arch 30, 2021 by guest. Protected by
http://adc.bmj.com
/A
rch Dis C
hild: first published as 10.1136/adc.39.204.125 on 1 April 1964.
D
ownloaded from
http://adc.bmj.com/
-
JAMES AND LASSMAN
. \ '' @' 1 ' ^An' t'^......"IFIG. 1.-Appearance at the end of
operation. The dura is open. The bone septum (Lv. 3) and the median
dura between the two dural tubeshave been removed. Diastematomyelia
(2 cm.) is in the centre; the bone septum had occupied the caudal
half and had been pressing on the
caudal spinal cord which is swollen and engorged (to the right
of the photograph) as compared with normal spinal cord (to the
left).
ossified the ventral attachment has been narrow andfrequently
more fibrous than bony, which makesavulsion of the septum from its
attachment deepdown very much easier. When the deep attachmentis
broad and osseous its removal is more difficult.It is a
misconception to regard a septum as growingfrom a vertebral body or
from a neural arch. Bonedevelops only in preformed tissues and
therefore it ispointless to discuss whether a septum
originatesdorsally or ventrally. The abnormality results
fromfailure of normal development in early embryoniclife and the
septum represents mature tissues formedfrom aberrant embryonic
cells.
Cases with Single Dural Tube. None of these 11cases had a septum
although one had a fatty fibrousplug between and adherent to the
two spinal cordsbut it was not adherent ventrally; this plug
wascontinuous dorsally through the dura to a largesubcutaneous
lipoma in the lumbar region. In everycase the bifid spinal cords
rejoined caudally to forman apparently normal spinal cord. In only
one case(not in the series) have we seen a condition that mighthave
been a bifid conus or duplicated filum terminale.There seems to be
no particular pathological signifi-cance in the extent of the
diastematomyelia. It hasvaried between 1 and 9 5 cm., although in
one casethe distance was not measured because the
diastematomyelia extended beyond the surgicalexposure. Where
there is diastematomyelia thetracts that normally cross from one
side to the othermust do so cranial or caudal to the division of
thespinal cord, but in a few of our cases there have
beenintervening bands that must be regarded as commis-sural.
Usually the space between the two spinal cords isquite obvious
but occasionally the two spinal cordsare very closely apposed,
their dorsal cleft beingmarked by a longitudinal dorsal blood
vessel in themidline. This fact has become evident in a
caseoperated on only recently and therefore not includedin this
series; such a finding, however, is unlikely tobe of any clinical
significance. There may havebeen other cases in which this was not
noticed atoperation.The associated extrinsic abnormality most
commonly found in this series was one or morebands that
microscopy has shown were sometimesnerves and sometimes fibrous
tissue. At one endthey were connected with one bifurcation of
thespinal cord and at the other end either to the dura orthrough
the dura to a neural arch (Table 1). Theyare almost always
angulated in their course andsince some cases have shown
improvement followingdivision of the offending band from its
attachment tothe dura, we can only postulate that they have
been
1216copyright.
on March 30, 2021 by guest. P
rotected byhttp://adc.bm
j.com/
Arch D
is Child: first published as 10.1136/adc.39.204.125 on 1 A
pril 1964. Dow
nloaded from
http://adc.bmj.com/
-
DIASTEMATOM YELIA 127
FIG. 2.-Appearance at operation after opening the dura mater
(single tube). An extradural band extended vertically from Lv. 2
neural archto attach to the dura and continued intrathecally in a
caudal direction (left of centre) to attach to the caudal junction
(centre) of the twospinal cords (seen on the left). The
diastematomyelia was 3 cm. long. Normal spinal cord to the right of
the exposure (Lv. 3 and 4 level)
FiG. 3.-Dura open and diastematomyelia 9.5 cm. long seen
throughout the exposure; neither bifurcation is shown. In the
centre is a nervepassing from the left to the right where it
penetrates the dura (Lv. 2 neural arch level). It receives
contributions from both spinal cordsand at its origin it lies over
and conceals commissural bands which connected through the dura to
the neural arch of Lv. 1; this connexioncan be seen left of centre
lying upon the surrounding swabs. There was a third similar band
from the neural arch of Tv. 12, which is not shown.
The two spinal cords were of equal size although the photograph
suggests that the left one was very large.
copyright. on M
arch 30, 2021 by guest. Protected by
http://adc.bmj.com
/A
rch Dis C
hild: first published as 10.1136/adc.39.204.125 on 1 April 1964.
D
ownloaded from
http://adc.bmj.com/
-
JAMES AND LASSMAN
exerting a traction effect. In some cases there hasbeen more
than one band, one attaching at eachbifurcation and one or more to
the commissuralbands between the two spinal cords, all having
adorsal attachment to the dorsal dura as well, oftencontinuing
extradurally to the neural arches (Figs. 2and 3).
TABLE 1
SINGLE DURAL TUBE-11 CASES: OPERATION FINDINGS
Operation Findings No. of Cases
Band connecting bifurcation of spinal cord with neuralarch . .
.. .. .. .. .. 5
Band connecting bifurcation of spinal cord with dorsaldura.
3
No bands but with filum terminale adherent to one sideand
rotating spinal cord . I
Lipomatous plug connecting to subcutaneous tissues.. 1No
extrinsic lesion affecting spinal cord 1
Total . .11
TABLE 2
11 CASES WITH SINGLE DURAL TUBE:RESULTS OF OPERATION
Before Operation No. of After OperationCases
Reflexes normal 2 Reflexes normal, no further deteri-oration in
gait or foot shape.
Reflexes abnormal 4 One or more reflexes changed tonormal.
3 Reflexes unchanged, gait improved.2 Reflexes unchanged, no
further
deterioration in gait or foot shape.
Total 11
The bands are aberrant posterior nerve rootseither still
functioning or atrophic. In the embryothe posterior nerve roots and
ganglia are formedfrom the cells of the neural crest; consequently
anyinterference with the development of the neural tubeis likely to
give rise to aberration in the migrations ofthe developing
cells.The single case in which no extrinsic lesion was
found was known, as the result of myelography, tohave two
lesions; the greater defect in the thoracicregion was explored and
the lesser defect in the lowerlumbar region was ignored. This was
an early caseand the first one in which the myelographic patternin
the lumbar area had been seen. Subsequent experi-ence has produced
several such cases and we nowknow that this child has another area
of diastemato-myelia with a band attaching to the dura and
possiblythe neural arch. The area has not been exploredsince the
child's clinical state has remained unaltered
in the two years following her thoracic laminectomy.We now
believe that in any case with two widelyseparated lesions it is
likely to be wiser to explore thelumbar region first. In theory the
clinical pictureshould indicate the level of the spinal cord which
isbeing affected, but since the conus medullaris israrely, if ever,
at the normal level in these cases, andthe pressure and traction
effects on the spinal cordare so diffuse, it is impossible to be
accurate inlocalizing the site of an extrinsic lesion.
In many of these cases it is difficult to know whythere should
be any neurological deficit and whyoperation should make any
difference. Table 2shows an analysis of the results of operation on
these11 cases in a follow-up of five months in one case,six months
in another and 14 months to three yearsin the remainder; but it is
not the purpose of thiscommunication to discuss the results of
operations,although they are encouraging.
Clinical FindingsThe syndrome of early neurological changes
occurring in spina bifida occulta, which we havedescribed, is
not the only mode of presentation anddevelopment. Spina bifida
occulta is a lesser degreeof spinal dysraphism than spina bifida
aperta(myelocele and meningomyelocele), but in somecases the
neurological deficit can become as severeand in exactly the same
way. Consequently, theappearance in a clinic of a child with lower
limbabnormalities or bladder or bowel disturbancesresembling those
found in spina bifida cystica shouldimmediately lead to the
investigation of the vertebralcolumn to make sure that there are no
laminaldefects. The clinical findings in diastematomyeliaare no
different from those in other cases of spinabifida occulta.
Table 3 shows an analysis of the symptoms withwhich our cases
presented. Pes cavovarus is theearliest evidence of progressive
neurological deficitand is usually unilateral to begin with. A
number ofchildren have some form of foot abnormality atbirth. Those
that start with club feet are commonlynot diagnosed as cases of
spinal dysraphism until alate stage and it is interesting to note
the number inTable 3. Those that start with calcaneous or flatfeet
and suddenly develop cavovarus are sufficientlydisconcerting to the
therapist for the diagnosis to bemade without much difficulty. If
sensory changesdevelop so that there is trophic ulceration,
acongenital club foot is clearly associated withneurological
deficit.
Pes valgus is a later stage of deterioration thancavovarus, and
the change from the latter is
I128
copyright. on M
arch 30, 2021 by guest. Protected by
http://adc.bmj.com
/A
rch Dis C
hild: first published as 10.1136/adc.39.204.125 on 1 April 1964.
D
ownloaded from
http://adc.bmj.com/
-
DIASTEMATOMYELIAunexpected and consequently striking. Where
thereis an external cutaneous manifestation on the back,e.g.
hypertrichosis, lipoma, naevus or dermal dimple,associated with
foot abnormality with or withoutreflex changes the diagnosis is
easier.Our single case with poor circulation in the legs
(James and Lassman, 1958) had the discoloration socommonly seen
in spina bifida cystica and in casesthat have suffered from
poliomyelitis in the past.
Table 4 is an analysis of the clinical findings in the24 cases
under discussion, and the only notable pointthat may be significant
is that there was no case withbowel or bladder disturbance; but
paraplegia isreported to have occurred in cases with a boneseptum.
In every other respect, Table 4 shows nofeature to distinguish
these cases from the other 36cases ofour series that did not have
diastematomyelia.
In the syndrome we have previously described, ashort leg and
foot was one of the basic factors andTable 4 shows how commonly it
is found. Of thefour cases in this group that were without
neurologicaldeficit, one had circulatory changes in the legs,
onehad a large subcutaneous lumbar lipoma, one hadweakness of the
leg, and one had hypertrichosis.The three with normal limbs had
marked externalmanifestations that warranted myelographic
investi-gation, and in every case an abnormality
wasdemonstrated.
Table 5 shows the cases that had external cuta-neous
manifestations on the back, usually in thelumbar or lumbosacral
region. These manifesta-tions occur randomly throughout our series
of 60cases, although it has been suggested by others
thathypertrichosis is an indication of underlyingdiastematomyelia;
this is not so.
Summary
In a series of 60 consecutive cases of spina bifidaocculta
submitted to laminectomy because ofabnormality of the spinal cord
or cauda equina,diastematomyelia was found in 24, and these
casesare surveyed.
Thirteen cases had separate dural tubes for eachspinal cord and
in every case there was a dorsi-ventral extradural septum as well;
11 had a singledural tube enclosing both spinal cords and none hada
septum between.The clinical appearances in these 24 cases of
diastematomyelia show no characteristic to distin-guish them
from the other 36 in the series.
It is the accompanying extrinsic lesion affectingthe spinal cord
that is of clinical importance and notthe diastematomyelia.
TABLE 3
PRESENTING SYMPTOMS IN 24 CASES
Symptoms No. of Cases
Pes cavovarus:Alone. 7With ulceration (3 cases relapsing talipes
equino-
varus) . 5Relapsing talipes equinovarus with cutaneous mani-
festation .. IDeveloping from valgus I
Total . .14
Pes valgus:Developing from cavovarus and ulcerated ..
2Developing from old talipes equinovarus 2
Total. 4
Cutaneous manifestations:Hypertrichosis. 2Lipoma a.Pigmented
hairy lipoma .. . . .
Total 4
Poor circulation in legs .Aching legs I
Total .24
TABLE 4
CLINICAL SYNDROME IN 24 CASES*
Clinical Syndrome No. ofCases
Normal apart from cutaneous manifestation 3Short leg, with
neurological deficit . . 3Short leg, without neurological deficit
I. . 1Short foot, without neurological deficit .. IShort foot and
leg, without neurological deficit .. 2Short foot and leg, with
neurological deficit .. I1Abnormal feet, with neurological deficit
.. 3
Total . .24
* None had bladder or bowel disturbance
TABLE 5CUTANEOUS MANIFESTATIONS ON THE BACK (24 CASES)
Cutaneous Manifestations No. of Cases
Dermal dimple:Alone .. .. .. .. .. .. .. 4With naevus 1
Lipoma.. 2Pigmented hairy lipoma .. .. .. ..Hypertrichosis:With
lipoma. 2With dermal dimple IWith naevus. 2Alone. 4
Total . 17
No cutaneous manifestation. . I 7
Total .24
129
copyright. on M
arch 30, 2021 by guest. Protected by
http://adc.bmj.com
/A
rch Dis C
hild: first published as 10.1136/adc.39.204.125 on 1 April 1964.
D
ownloaded from
http://adc.bmj.com/
-
130 JAMES AND LASSMANWe are glad to acknowledge our continued
indebtedness
to Dr. Gordon Gryspeerdt, Neuroradiologist, Dr.Douglas Whitby,
Anaesthetist, and Sister C. Palfreyman,Ward Sister, who have
collaborated closely with us inthis work.Mr. R. W. Ridley and Mrs.
P. Bone of the University
Department of Medical Photography have photographedall our
cases.The Scientific and Research Committee of the
Newcastle Regional Hospital Board has providedassistance and
finance for morbid anatomical investiga-tions related to this
work.
REFERENCES
Gryspeerdt, G. L. (1963). Myelographic assessment of occult
formsof spinal dysraphism. Acta radiol. (Stockh.), n.s., 1,
702.
James, C. C. M. and Lassman, L. P. (1958).
Diastematomyelia.Arch. Dis. Childh., 33, 536.- (1960). Spinal
dysraphism. An orthopaedic syndrome inchildren accompanying occult
forms. ibid., 35, 315.-(1962a). Spinal dysraphism. The diagnosis
and treatmentof progressive lesions in spina bifida occulta. J.
Bone Jt Surg.,44B, 828.- --(1962b). Spinal dysraphism. Spinal cord
lesions associa-
ted with spina bifida occulta. Physiotherapy, 48, 154.Lassman,
L. P. and James, C. C. M. (1963). Lumbosacral lipomata
and lesions of the conus medullaris and cauda equina.
Excerptamed. (Amst.), Int. Cong. Series, No. 60, p. 139.
copyright. on M
arch 30, 2021 by guest. Protected by
http://adc.bmj.com
/A
rch Dis C
hild: first published as 10.1136/adc.39.204.125 on 1 April 1964.
D
ownloaded from
http://adc.bmj.com/
